计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (14): 25-27.

• 研究、探讨 • 上一篇    下一篇

参数最优化的GM(1,1)模型

李 蔚   

  1. 湘潭大学 能源工程学院,湖南 湘潭 411105
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-05-11 发布日期:2011-05-11

GM(1,1) model with optimal parameters

LI Wei   

  1. Energy Engineering College,Xiangtan University,Xiangtan,Hunan 411105,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-05-11 Published:2011-05-11

摘要: 灰色GM(1,1)模型的精度取决于参数[a]和[u],而参数[a]和[u]的获得依赖于初始条件的选择和背景值的构造。在背景值的计算中引入优化因子,通过最优初始条件和最佳优化因子的选择使预测结果与原始数据的平均相对误差最小,从而得到参数最优化的GM(1,1)模型。数值试验的结果表明,相对于传统的GM(1,1)模型及文献中的改进方法,参数最优化的GM(1,1)模型具有更高的预测精度。

关键词: GM(1, 1)模型, 精度, 优化因子, 初始条件

Abstract: The precision of GM(1,1) model is decided by the parameters [a] and [u],while the parameters [a] and [u]depend on the selection of original condition and the conformation of background values.Optimization factor is introduced into the conformation of background values,and the new GM(1,1) model with optimal parameters is obtained by choosing the optimal original condition and optimization factor.Numerical tests indicate that the GM(1,1) model with optimal parameters has higher precision than the traditional GM(1,1) model and the new GM(1,1) model in the literature.

Key words: GM(1, 1) model, precision, optimization factor, original condition